Computational Image Analysis

Objectives

Long Term goals are the in-depth understanding on processing medical images, and the development of algorithms capable to process huge amount of data, fully automatic and being applicable to different imaging resolutions. For this reason we will concentrate on novel state-of-the art image acquisition devices, such as optical coherence tomography (OCT) and scanning laser optical projection tomography (OPT/SLOT). Additionally, we will concentrate on different semantic levels of image interpretation, ranging from image segmentation frameworks, e.g. using Markov Random Fields, up to Boosting methods for detection or registration.

In the next years, the research group will further bridge between medicine and engineering. Due to the recent developments in big data analysis and machine learning, we will continue to use these methods in open questions in medical data science. Automatic analysis tools ranging from unsupervised learning, clustering, up to semi-supervised and supervised deep learning are the methods of choice to invent novel strategies for biomarker or therapeutic research.

Milestones: 

  • Better understanding on OCT imaging, refractive index computation and motion compensation 
  • Image registration and content analysis for 3D structured data sets to obtain a statistical representation of organic structures 
  • Generation of a morphable organ model, which allows for the individualized synthesis of organ components. 

Research Focus

Regenerative processes are complex and multicellular dynamic events, resulting from an orchestrated interaction at different levels, ranging from a subcellular range to physiology up to organ scale. Being able to trace, document and visualize parts of the regeneration process is therefore essential for the study, development and control of these events. It requires multiple imaging platforms providing both, standard and new imaging techniques to satisfy respective needs of the cluster. 

Based on novel state-of-the art image acquisition devices, such as optical coherence tomography (OCT) and scanning laser optical projection tomography (OPT/SLOT) novel computer vision methods will be developed to segment and reconstruct relevant structures (e.g. cell locations, orientations, variations, ligament densities, vessel branches, etc.) from such 3D input data. An additional focus will be the development of time-predictive algorithms to control and steer organ growth from sample data (e.g. using Hidden Markov Models, HMM). This will require a detailed analysis and correlation of 3D structured data over time, e.g. using methods from multivariate statistics (e.g. based on a principal component analysis, PCA). Another application will be a statistical model for organ structures, which can be used to design individual organs for respective patients, e.g. by integration of information about height, weight, age in a modified subspace projection approach or 3D texture synthesis methods. 

Collaborations

Publications

2013 - ongoing

2017

Buch A, Müller O, Ivanova L, Döhner K, Bialy D, Bosse JB, Pohlmann A, Binz A, Hegemann M, Nagel C-H, Koltzenburg M, Viejo-Borbolla A, Rosenhahn B, Bauerfeind R, Sodeik B. Inner tegument proteins of Herpes Simplex Virus are sufficient for intracellular capsid motility in neurons but not for axonal targeting. PLOS Pathogens. 2017;13(12):e1006813.

Kluger F, Ackermann H, Yang MY, Rosenhahn B. Deep Learning for Vanishing Point Detection Using an Inverse Gnomonic Projection. In: Roth V, Vetter T, editors. Pattern Recognition: 39th German Conference, GCPR 2017, Basel, Switzerland, September 12–15, 2017, Proceedings. Cham: Springer International Publishing; 2017. p. 17-28.

Müller O, Rosenhahn B, editors. Global Consistency Priors for Joint Part-Based Object Tracking and Image Segmentation. 2017 IEEE Winter Conference on Applications of Computer Vision (WACV); 2017 24-31 March 2017.

Zell P, Wandt B, Rosenhahn B, editors. Joint 3D Human Motion Capture and Physical Analysis From Monocular Videos. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops; 2017.

Zell P, Rosenhahn B, editors. Learning-Based Inverse Dynamics of Human Motion. The IEEE International Conference on Computer Vision (ICCV); 2017.

Yang MY, Ackermann H, Lin WY, Feng ST, Rosenhahn B. Motion Segmentation Using Global and Local Sparse Subspace Optimization. Photogrammetric Engineering and Remote Sensing. 2017;83(11):769-78.

von Marcard T, Rosenhahn B, Black MJ, Pons-Moll G. Sparse Inertial Poser: Automatic 3D Human Pose Estimation from Sparse IMUs. Computer Graphics Forum. 2017;36(2):349-60.

Liao W, Yang C, Yang MY, Rosenhahn B. Security Event Recognition for Visual Surveillance. ISPRS Annals of Photogrammetry, Remote Sensing & Spatial Information Sciences. 2017;4.

Bushnevskiy A, Sorgi L, Rosenhahn B. Feature Points Densification and Refinement.  19th International Conference on Image Analysis and Processing (ICIAP); 2017; Italy2017.

Berthold T, Leichter A, Rosenhahn B, Berkhahn V, Valerius J, editors. Seabed Sediment Classification of Side-scan Sonar Data Using Convolutional Neural Networks IEEE Symposium Series on Computational Intelligence 2017; BerLei2017.

Alldieck T, Kassubeck M, Wandt B, Rosenhahn B, Magnor M. Optical Flow-Based 3D Human Motion Estimation from Monocular Video. In: Roth V, Vetter T, editors. Pattern Recognition: 39th German Conference, GCPR 2017, Basel, Switzerland, September 12–15, 2017, Proceedings. Cham: Springer International Publishing; 2017. p. 347-60.

2016

Yang M, Rosenhahn B. Superpixel cut for gure-ground image segmentation.  SPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences; 2016.

Zell P, Wandt B, Rosenhahn B. Physics-based Models for Human Gait Analysis. In: Müller BW, Sebastian I., editor. Physics-based Models for Human Gait Analysis: Springer International Publishing; 2016.

Wandt B, Ackermann H, Rosenhahn B. 3D Reconstruction of Human Motion from Monocular Image Sequences. IEEE Trans Pattern Anal Mach Intell. 2016;38(8):1505-16.

von Marcard T, Pons-Moll G, Rosenhahn B. Human Pose Estimation from Video and IMUs. IEEE Trans Pattern Anal Mach Intell. 2016.

Schlobohm J, Pösch A, Reithmeier E, Rosenhahn B. Improving contour based pose estimation for fast 3D measurement of free form objects. Measurement. 2016;92:79-82.

Perez-Pellitero E, Salvador J, Ruiz-Hidalgo J, Rosenhahn B. Antipodally Invariant Metrics for Fast Regression-Based Super-Resolution. IEEE Trans Image Process. 2016;25(6):2456-68.

2015

Zell P, Rosenhahn B. A Physics-Based Statistical Model for Human Gait Analysis. In: Gall J, Gehler P, Leibe B, editors. Pattern Recognition: 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings. Cham: Springer International Publishing; 2015. p. 169-80.

Salman Al-Shaikhli SD, Yang MY, Rosenhahn B. 3d Automatic Liver Segmentation Using Feature-Constrained Mahalanobis Distance in Ct Images. Biomed Tech (Berl). 2015. Epub 2015/10/27.

Salman Al-Shaikhli Saif D, Yang Michael Y, Rosenhahn B. Brain Tumor Classification and Segmentation Using Sparse Coding and Dictionary Learning. Biomedical Engineering / Biomedizinische Technik2015.

Salman Al-shaikhli Saif D, Yang Michael Y, Rosenhahn B. 3d Automatic Liver Segmentation Using Feature-Constrained Mahalanobis Distance in Ct Images. Biomedical Engineering / Biomedizinische Technik2015.

Kuznetsova A, Sung Ju H, Rosenhahn B, Sigal L, editors. Expanding Object Detector's Horizon: Incremental Learning Framework for Object Detection in Videos. Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on; 2015 7-12 June 2015.

2014

Shoaib M, Yang MY, Rosenhahn B, Ostermann J. Estimating Layout of Cluttered Indoor Scenes Using Trajectory-Based Priors. Image and Vision Computing. 2014;32(11):870-83.

Henseler H, Kuznetsova A, Vogt P, Rosenhahn B. Validation of the Kinect Device as a New Portable Imaging System for Three-Dimensional Breast Assessment. J Plast Reconstr Aesthet Surg. 2014;67(4):483-8.

Bartsch I, Willbold E, Rosenhahn B, Witte F. Non-Invasive Ph Determination Adjacent to Degradable Biomaterials in Vivo. Acta Biomater. 2014;10(1):34-9.

Al-Shaikhli SD, Yang MY, Rosenhahn B. Multi-Region Labeling and Segmentation Using a Graph Topology Prior and Atlas Information in Brain Images. Comput Med Imaging Graph. 2014;38(8):725-34.

2013

Kuznetsova A, Leal-Taixé L, Rosenhahn B. Real-time sign language recognition using a consumer depth camera. IEEE International Conference on Computer Vision Workshops (ICCVW), 3rd Workshop on Consumer Depth Cameras for Computer Vision (CDC4CV), December 2013.

Scheuermann B, Gkoutelitsas S, Rosenhahn B. Multi-Sensor Fusion Using Dempster’s Theory of Evidence for Video Segmentation. In: Ruiz-Shulcloper J, Sanniti di Baja G, editors. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications: Springer Berlin Heidelberg; 2013. p. 431-8.

Muller O, Yang MY, Rosenhahn B, editors. Slice Sampling Particle Belief Propagation. Computer Vision (ICCV), 2013 IEEE International Conference on; 2013 1-8 Dec. 2013.

Donner S, Muller O, Witte F, Bartsch I, Willbold E, Ripken T, Heisterkamp A, Rosenhahn B, Kruger A. In Situ Optical Coherence Tomography of Percutaneous Implant-Tissue Interfaces in a Murine Model. Biomed Tech (Berl). 2013;58(4):359-67.

2006 - 2012

2012

Spindler R, Rosenhahn B, Hofmann N, Glasmacher B. Video Analysis of Osmotic Cell Response During Cryopreservation. Cryobiology. 2012;64(3):250-60.

Muller O, Donner S, Klinder T, Bartsch I, Kruger A, Heisterkamp A, Rosenhahn B. Compensating Motion Artifacts of 3d in Vivo Sd-Oct Scans. Med Image Comput Comput Assist Interv. 2012;15(Pt 1):198-205.

Kuznetsova A, Pons-Moll G, Rosenhahn B. Pca-Enhanced Stochastic Optimization Methods. In: Pinz A, Pock T, Bischof H, Leberl F, editors. Pattern Recognition: Springer Berlin Heidelberg; 2012. p. 377-86.

2011

Muller O, Donner S, Klinder T, Dragon R, Bartsch I, Witte F, Kruger A, Heisterkamp A, Rosenhahn B. Model Based 3d Segmentation and Oct Image Undistortion of Percutaneous Implants. Med Image Comput Comput Assist Interv. 2011;14(Pt 3):454-62.

Ehlers A, Baumann F, Spindler R, Glasmacher B, Rosenhahn B. Pca Enhanced Training Data for Adaboost. In: Real P, Diaz-Pernil D, Molina-Abril H, Berciano A, Kropatsch W, editors. Computer Analysis of Images and Patterns: Springer Berlin Heidelberg; 2011. p. 410-9.

Helga H, Kim BS, Peter Maria V, Bodo R. The Kinect Recording System for Objective Three- and Four-Dimensional Breast Assessment with Image Overlays. Journal of Plastic, Reconstructive & Aesthetic Surgery.

Al-Shaikhli SDS, Yang MY, Rosenhahn B. Multi-Region Labeling and Segmentation Using a Graph Topology Prior and Atlas Information in Brain Images. Computerized Medical Imaging and Graphics.38(8):725-34.

test
Diese Webseite verwendet Cookies und die Webanalyse-Software Google Analytics. Klicken Sie hier für weitere Informationen zum Datenschutz.
Einverstanden